From: Jan Blunck <jblunck@infradead.org>
To: dev@dpdk.org
Subject: [dpdk-dev] [PATCH 09/38] net/af_packet: use ethdev allocation helper for virtual devices
Date: Mon, 6 Mar 2017 11:00:01 +0100 [thread overview]
Message-ID: <1488794430-25179-10-git-send-email-jblunck@infradead.org> (raw)
In-Reply-To: <1488794430-25179-1-git-send-email-jblunck@infradead.org>
Signed-off-by: Jan Blunck <jblunck@infradead.org>
---
drivers/net/af_packet/rte_eth_af_packet.c | 42 ++++++++++++++-----------------
1 file changed, 19 insertions(+), 23 deletions(-)
diff --git a/drivers/net/af_packet/rte_eth_af_packet.c b/drivers/net/af_packet/rte_eth_af_packet.c
index 77536e8..6f6ba0c 100644
--- a/drivers/net/af_packet/rte_eth_af_packet.c
+++ b/drivers/net/af_packet/rte_eth_af_packet.c
@@ -38,6 +38,7 @@
#include <rte_mbuf.h>
#include <rte_ethdev.h>
+#include <rte_ethdev_vdev.h>
#include <rte_malloc.h>
#include <rte_kvargs.h>
#include <rte_vdev.h>
@@ -539,18 +540,19 @@ open_packet_iface(const char *key __rte_unused,
static struct rte_vdev_driver pmd_af_packet_drv;
static int
-rte_pmd_init_internals(const char *name,
+rte_pmd_init_internals(struct rte_vdev_device *dev,
const int sockfd,
const unsigned nb_queues,
unsigned int blocksize,
unsigned int blockcnt,
unsigned int framesize,
unsigned int framecnt,
- const unsigned numa_node,
struct pmd_internals **internals,
struct rte_eth_dev **eth_dev,
struct rte_kvargs *kvlist)
{
+ const char *name = rte_vdev_device_name(dev);
+ const unsigned int numa_node = dev->device.numa_node;
struct rte_eth_dev_data *data = NULL;
struct rte_kvargs_pair *pair = NULL;
struct ifreq ifr;
@@ -768,7 +770,7 @@ rte_pmd_init_internals(const char *name,
}
/* reserve an ethdev entry */
- *eth_dev = rte_eth_dev_allocate(name);
+ *eth_dev = rte_eth_vdev_allocate(dev, 0);
if (*eth_dev == NULL)
goto error;
@@ -782,22 +784,16 @@ rte_pmd_init_internals(const char *name,
(*internals)->nb_queues = nb_queues;
+ rte_memcpy(data, (*eth_dev)->data, sizeof(*data));
data->dev_private = *internals;
- data->port_id = (*eth_dev)->data->port_id;
data->nb_rx_queues = (uint16_t)nb_queues;
data->nb_tx_queues = (uint16_t)nb_queues;
data->dev_link = pmd_link;
data->mac_addrs = &(*internals)->eth_addr;
- strncpy(data->name,
- (*eth_dev)->data->name, strlen((*eth_dev)->data->name));
(*eth_dev)->data = data;
(*eth_dev)->dev_ops = &ops;
- (*eth_dev)->driver = NULL;
(*eth_dev)->data->dev_flags = RTE_ETH_DEV_DETACHABLE;
- (*eth_dev)->data->drv_name = pmd_af_packet_drv.driver.name;
- (*eth_dev)->data->kdrv = RTE_KDRV_NONE;
- (*eth_dev)->data->numa_node = numa_node;
return 0;
@@ -822,11 +818,11 @@ rte_pmd_init_internals(const char *name,
}
static int
-rte_eth_from_packet(const char *name,
+rte_eth_from_packet(struct rte_vdev_device *dev,
int const *sockfd,
- const unsigned numa_node,
struct rte_kvargs *kvlist)
{
+ const char *name = rte_vdev_device_name(dev);
struct pmd_internals *internals = NULL;
struct rte_eth_dev *eth_dev = NULL;
struct rte_kvargs_pair *pair = NULL;
@@ -909,11 +905,11 @@ rte_eth_from_packet(const char *name,
RTE_LOG(INFO, PMD, "%s:\tframe size %d\n", name, framesize);
RTE_LOG(INFO, PMD, "%s:\tframe count %d\n", name, framecount);
- if (rte_pmd_init_internals(name, *sockfd, qpairs,
- blocksize, blockcount,
- framesize, framecount,
- numa_node, &internals, ð_dev,
- kvlist) < 0)
+ if (rte_pmd_init_internals(dev, *sockfd, qpairs,
+ blocksize, blockcount,
+ framesize, framecount,
+ &internals, ð_dev,
+ kvlist) < 0)
return -1;
eth_dev->rx_pkt_burst = eth_af_packet_rx;
@@ -925,15 +921,12 @@ rte_eth_from_packet(const char *name,
static int
rte_pmd_af_packet_probe(struct rte_vdev_device *dev)
{
- const char *name = rte_vdev_device_name(dev);
- unsigned numa_node;
int ret = 0;
struct rte_kvargs *kvlist;
int sockfd = -1;
- RTE_LOG(INFO, PMD, "Initializing pmd_af_packet for %s\n", name);
-
- numa_node = rte_socket_id();
+ RTE_LOG(INFO, PMD, "Initializing pmd_af_packet for %s\n",
+ rte_vdev_device_name(dev));
kvlist = rte_kvargs_parse(rte_vdev_device_args(dev), valid_arguments);
if (kvlist == NULL) {
@@ -953,7 +946,10 @@ rte_pmd_af_packet_probe(struct rte_vdev_device *dev)
goto exit;
}
- ret = rte_eth_from_packet(name, &sockfd, numa_node, kvlist);
+ if (dev->device.numa_node == SOCKET_ID_ANY)
+ dev->device.numa_node = rte_socket_id();
+
+ ret = rte_eth_from_packet(dev, &sockfd, kvlist);
close(sockfd); /* no longer needed */
exit:
--
2.7.4
next prev parent reply other threads:[~2017-03-06 10:00 UTC|newest]
Thread overview: 104+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-03-06 9:59 [dpdk-dev] [PATCH 00/38] Remove struct eth_driver Jan Blunck
2017-03-06 9:59 ` [dpdk-dev] [PATCH 01/38] eal: add name field to generic device Jan Blunck
2017-03-06 9:59 ` [dpdk-dev] [PATCH 02/38] eal: parse "driver" device argument before probing drivers Jan Blunck
2017-03-06 9:59 ` [dpdk-dev] [PATCH 03/38] net/nfp: use library function for DMA zone reserve Jan Blunck
2017-03-10 7:03 ` Shreyansh Jain
2017-03-10 7:20 ` Jan Blunck
2017-03-06 9:59 ` [dpdk-dev] [PATCH 04/38] net/vmxnet3: " Jan Blunck
2017-03-06 9:59 ` [dpdk-dev] [PATCH 05/38] ether: add allocation helper for virtual drivers Jan Blunck
2017-03-06 9:59 ` [dpdk-dev] [PATCH 06/38] net/tap: use ethdev allocation helper for virtual devices Jan Blunck
2017-03-06 9:59 ` [dpdk-dev] [PATCH 07/38] net/vhost: " Jan Blunck
2017-03-06 10:00 ` [dpdk-dev] [PATCH 08/38] net/virtio: " Jan Blunck
2017-03-06 10:00 ` Jan Blunck [this message]
2017-03-06 10:00 ` [dpdk-dev] [PATCH 10/38] app/test: don't short-circuit null device creation Jan Blunck
2017-03-06 10:00 ` [dpdk-dev] [PATCH 11/38] net/null: internalize eth_dev_null_create() Jan Blunck
2017-03-06 10:00 ` [dpdk-dev] [PATCH 12/38] net/null: use ethdev allocation helper for virtual devices Jan Blunck
2017-03-06 10:00 ` [dpdk-dev] [PATCH 13/38] net/bonding: make bonding API call through EAL on create/free Jan Blunck
2017-03-06 10:00 ` [dpdk-dev] [PATCH 14/38] net/bonding: use ethdev allocation helper for virtual devices Jan Blunck
2017-03-06 10:00 ` [dpdk-dev] [PATCH 15/38] ethdev: add PCI driver helpers Jan Blunck
2017-03-06 10:00 ` [dpdk-dev] [PATCH 16/38] net/virtio: Don't use eth_driver Jan Blunck
2017-03-06 10:00 ` [dpdk-dev] [PATCH 17/38] net/bnx2x: " Jan Blunck
2017-03-06 19:32 ` Harish Patil
2017-03-06 10:00 ` [dpdk-dev] [PATCH 18/38] net/bnxt: " Jan Blunck
2017-03-06 10:00 ` [dpdk-dev] [PATCH 19/38] net/cxgbe: " Jan Blunck
2017-03-06 10:00 ` [dpdk-dev] [PATCH 20/38] net/em: " Jan Blunck
2017-03-06 10:00 ` [dpdk-dev] [PATCH 21/38] net/igb: " Jan Blunck
2017-03-06 10:00 ` [dpdk-dev] [PATCH 22/38] net/ena: " Jan Blunck
2017-03-06 10:00 ` [dpdk-dev] [PATCH 23/38] net/enic: " Jan Blunck
2017-03-06 10:00 ` [dpdk-dev] [PATCH 24/38] net/fm10k: " Jan Blunck
2017-03-06 10:00 ` [dpdk-dev] [PATCH 25/38] net/i40e: " Jan Blunck
2017-03-06 10:00 ` [dpdk-dev] [PATCH 26/38] net/i40evf: " Jan Blunck
2017-03-06 10:00 ` [dpdk-dev] [PATCH 27/38] net/ixgbe: " Jan Blunck
2017-03-06 10:00 ` [dpdk-dev] [PATCH 28/38] net/mlx: Don't reference eth_driver Jan Blunck
2017-03-06 10:00 ` [dpdk-dev] [PATCH 29/38] net/nfp: Don't use eth_driver Jan Blunck
2017-03-06 10:00 ` [dpdk-dev] [PATCH 30/38] net/qede: " Jan Blunck
2017-03-06 19:33 ` Harish Patil
2017-03-06 10:00 ` [dpdk-dev] [PATCH 31/38] net/sfc: " Jan Blunck
2017-03-06 10:00 ` [dpdk-dev] [PATCH 32/38] net/szedata2: " Jan Blunck
2017-03-06 10:00 ` [dpdk-dev] [PATCH 33/38] net/thunderx: " Jan Blunck
2017-03-06 10:00 ` [dpdk-dev] [PATCH 34/38] net/vmxnet3: " Jan Blunck
2017-03-06 10:00 ` [dpdk-dev] [PATCH 35/38] ethdev: remove unused ethdev PCI probe/remove Jan Blunck
2017-03-06 10:00 ` [dpdk-dev] [PATCH 36/38] ethdev: remove unused ethdev driver Jan Blunck
2017-03-06 10:00 ` [dpdk-dev] [PATCH 37/38] ethdev: remove PCI specific helper from generic ethdev header Jan Blunck
2017-03-06 10:00 ` [dpdk-dev] [PATCH 38/38] ethdev: don't include PCI header Jan Blunck
2017-03-23 15:34 ` [dpdk-dev] [PATCH 00/38] Remove struct eth_driver Stephen Hemminger
2017-03-25 10:50 ` Jan Blunck
2017-04-11 15:44 ` [dpdk-dev] [PATCH v2 00/42] " Gaetan Rivet
2017-04-11 15:44 ` [dpdk-dev] [PATCH v2 01/42] eal: add name field to generic device Gaetan Rivet
2017-04-11 15:44 ` [dpdk-dev] [PATCH v2 02/42] eal: parse "driver" device argument before probing drivers Gaetan Rivet
2017-05-10 2:34 ` Ferruh Yigit
2017-05-10 11:01 ` Ferruh Yigit
2017-04-11 15:44 ` [dpdk-dev] [PATCH v2 03/42] net/nfp: use library function for DMA zone reserve Gaetan Rivet
2017-04-11 15:44 ` [dpdk-dev] [PATCH v2 04/42] ether: add allocation helper for virtual drivers Gaetan Rivet
2017-04-11 15:44 ` [dpdk-dev] [PATCH v2 05/42] net/tap: use ethdev allocation helper for virtual devices Gaetan Rivet
2017-04-11 15:44 ` [dpdk-dev] [PATCH v2 06/42] net/vhost: " Gaetan Rivet
2017-04-11 15:44 ` [dpdk-dev] [PATCH v2 07/42] net/virtio: " Gaetan Rivet
2017-04-11 15:44 ` [dpdk-dev] [PATCH v2 08/42] net/af_packet: " Gaetan Rivet
2017-04-11 15:44 ` [dpdk-dev] [PATCH v2 09/42] app/test: don't short-circuit null device creation Gaetan Rivet
2017-04-11 15:44 ` [dpdk-dev] [PATCH v2 10/42] net/null: internalize eth_dev_null_create() Gaetan Rivet
2017-04-11 15:44 ` [dpdk-dev] [PATCH v2 11/42] net/null: use ethdev allocation helper for virtual devices Gaetan Rivet
2017-04-11 15:44 ` [dpdk-dev] [PATCH v2 12/42] net/bonding: make bonding API call through EAL on create/free Gaetan Rivet
2017-04-11 15:44 ` [dpdk-dev] [PATCH v2 13/42] net/bonding: use ethdev allocation helper for virtual devices Gaetan Rivet
2017-04-11 15:44 ` [dpdk-dev] [PATCH v2 14/42] net/kni: " Gaetan Rivet
2017-04-11 15:44 ` [dpdk-dev] [PATCH v2 15/42] net/pcap: " Gaetan Rivet
2017-04-11 15:44 ` [dpdk-dev] [PATCH v2 16/42] ethdev: add PCI driver helpers Gaetan Rivet
2017-04-11 15:44 ` [dpdk-dev] [PATCH v2 17/42] net/virtio: Don't use eth_driver Gaetan Rivet
2017-04-11 15:44 ` [dpdk-dev] [PATCH v2 18/42] net/bnx2x: " Gaetan Rivet
2017-04-11 15:44 ` [dpdk-dev] [PATCH v2 19/42] net/bnxt: " Gaetan Rivet
2017-04-11 15:44 ` [dpdk-dev] [PATCH v2 20/42] net/cxgbe: " Gaetan Rivet
2017-04-11 15:44 ` [dpdk-dev] [PATCH v2 21/42] net/em: " Gaetan Rivet
2017-04-11 15:44 ` [dpdk-dev] [PATCH v2 22/42] net/igb: " Gaetan Rivet
2017-04-11 15:44 ` [dpdk-dev] [PATCH v2 23/42] net/ena: " Gaetan Rivet
2017-04-11 15:44 ` [dpdk-dev] [PATCH v2 24/42] net/enic: " Gaetan Rivet
2017-04-11 15:44 ` [dpdk-dev] [PATCH v2 25/42] net/fm10k: " Gaetan Rivet
2017-04-11 15:44 ` [dpdk-dev] [PATCH v2 26/42] net/i40e: " Gaetan Rivet
2017-04-11 15:44 ` [dpdk-dev] [PATCH v2 27/42] net/i40evf: " Gaetan Rivet
2017-04-11 15:44 ` [dpdk-dev] [PATCH v2 28/42] net/ixgbe: " Gaetan Rivet
2017-04-11 15:44 ` [dpdk-dev] [PATCH v2 29/42] net/mlx: Don't reference eth_driver Gaetan Rivet
2017-04-11 15:44 ` [dpdk-dev] [PATCH v2 30/42] net/nfp: Don't use eth_driver Gaetan Rivet
2017-04-11 15:44 ` [dpdk-dev] [PATCH v2 31/42] net/qede: " Gaetan Rivet
2017-04-11 15:44 ` [dpdk-dev] [PATCH v2 32/42] net/sfc: " Gaetan Rivet
2017-04-11 15:44 ` [dpdk-dev] [PATCH v2 33/42] net/szedata2: " Gaetan Rivet
2017-04-11 15:44 ` [dpdk-dev] [PATCH v2 34/42] net/thunderx: " Gaetan Rivet
2017-04-11 15:44 ` [dpdk-dev] [PATCH v2 35/42] net/vmxnet3: " Gaetan Rivet
2017-04-11 15:44 ` [dpdk-dev] [PATCH v2 36/42] net/avp: " Gaetan Rivet
2017-04-11 18:48 ` Legacy, Allain
2017-04-11 15:44 ` [dpdk-dev] [PATCH v2 37/42] net/liquidio: " Gaetan Rivet
2017-04-11 15:44 ` [dpdk-dev] [PATCH v2 38/42] ethdev: remove unused ethdev PCI probe/remove Gaetan Rivet
2017-04-12 11:24 ` Neil Horman
2017-04-12 11:28 ` Neil Horman
2017-04-11 15:44 ` [dpdk-dev] [PATCH v2 39/42] ethdev: remove unused ethdev driver Gaetan Rivet
2017-04-11 15:44 ` [dpdk-dev] [PATCH v2 40/42] test: " Gaetan Rivet
2017-04-11 15:44 ` [dpdk-dev] [PATCH v2 41/42] ethdev: remove PCI specific helper from generic ethdev header Gaetan Rivet
2017-04-11 15:44 ` [dpdk-dev] [PATCH v2 42/42] ethdev: don't include PCI header Gaetan Rivet
2017-04-12 16:25 ` [dpdk-dev] [PATCH v2 00/42] Remove struct eth_driver Stephen Hemminger
2017-04-14 13:09 ` Thomas Monjalon
2017-04-18 18:27 ` [dpdk-dev] [PATCH 0/2] next-net: remove ethdev driver Ferruh Yigit
2017-04-18 18:27 ` [dpdk-dev] [PATCH 1/2] net/ark: remove eth_dev Ferruh Yigit
2017-04-18 19:24 ` Ed Czeck
2017-04-19 10:03 ` Ferruh Yigit
2017-04-18 18:27 ` [dpdk-dev] [PATCH 2/2] net/xenvirt: remove ethdev driver Ferruh Yigit
2017-04-19 9:52 ` [dpdk-dev] [PATCH v2] net/xenvirt: fix build error Ferruh Yigit
2017-04-19 12:12 ` Thomas Monjalon
2017-04-18 18:38 ` [dpdk-dev] [PATCH 0/2] next-net: remove ethdev driver Ferruh Yigit
2017-04-19 11:37 ` Shreyansh Jain
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=1488794430-25179-10-git-send-email-jblunck@infradead.org \
--to=jblunck@infradead.org \
--cc=dev@dpdk.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).